Inferring in Vivo Capsid Assembly Kinetics from in Vitro by Stochastic Simulation The proposed work will use computer simulations to predict how differences between in vitro models and the in vivo cellular environment can be expected to affect assembly pathways of viral capsids. Viral capsids are one of the key model systems for understanding complex self- assembly in biology. Detailed information about their assembly process derives predominantly from in vitro models because of the infeasibility of examining the assembly process in living cells. Yet the chemical environment of a cell differs from that of these in vitro model systems in several respects known to be particularly important for macromolecular assembly reactions, including much higher local concentrations, dense macromolecular crowding, presence of viral genomes, and stochastic effects due to small reactant copy numbers. Experimental and theoretical results suggest that such changes can dramatically affect assembly rates and even basic choices of assembly pathways, raising the question of how reliable in vitro assembly data are and whether we can better interpret them to predict likely assembly behavior in vivo. The proposed work will use stochastic simulations designed to model biochemistry at cellular scales to test the hypothesis that changes from an in vitro to an in vivo environment will alter favored capsid assembly pathways. The work will examine three model systems for which in vitro assembly data are available: hepatitis B virus (HBV), cowpea chlorotic mottle virus (CCMV), and human papillomavirus (HPV). These simulations will then be translated in silico from in vitro conditions to conditions better resembling the in vivo environment, including high viral protein concentrations, densely crowded media, and the presence of nucleic acid. For each condition, a range of likely assembly pathways will be assessed to determine whether the change in environment is sufficient to alter the overall assembly mechanism. The work will have several implications for human health, as well as basic biophysical research. Capsid assembly has become a potential target of anti-viral drugs, creating a need for identification of critical points at which the assembly process is most vulnerable. The work will lead to general strategies for better inferring in vivo reaction mechanisms from in vitro data and will provide specific guidance for two human pathogens: HBV and HPV. It will further provide guidance on how both simulation methods and in vitro assembly models might be improved to better represent biochemistry in vivo. PUBLIC HEALTH RELEVANCE: Inferring in Vivo Capsid Assembly Kinetics from in Vitro by Stochastic Simulation The proposed work will affect human health by identifying likely key steps in assembly of two viruses implicated in human disease: hepatitis B virus and papillomavirus. This information is valuable for designing anti-viral agents targeted to the capsid assembly process. The work will also have broader relevance to understanding virus assembly in general and to developing computer models useful in a broad range of biomedical modeling applications.